Single-cell RNA-seq analysis of SARS-CoV-2 infection

The raw sequencing reads were demultiplexed using Cell Ranger mkfastq (10x Genomics). We trimmed the reads from the BIDMC liver samples for polyA tails and the template switching oligo 5’-AAGCAGTGGTATCAACGCAGAGTACATrGrGrG −3’ with cutadapt v.2.7⁷¹ . The reads were aligned to generate the count matrix using Cell Ranger count (10x Genomics) on Terra with the cellranger_workflow in Cumulus^{72 (link)}. The reads were aligned to a custom-built Human GRCh38 and SARS-CoV-2 (“GRCh38_premrna_and_SARSCoV2”) RNA reference. The GRCh38 pre-mrna reference captures reads mapping to both exons or introns^{73 (link)}. The SARS-CoV-2 viral sequence (FASTA file) and accompanying gene annotation and structure (GTF file) are as previously described^{74 (link)}. The GTF file was edited to include only CDS regions, with added regions for the 5’ UTR (“SARSCoV2_5prime”), 3’ UTR (“SARSCoV2_3prime”), and anywhere within the Negative Strand (“SARSCoV2_NegStrand”) of SARS-CoV-2. Trailing A’s at the 3’ end of the virus were excluded from the SARSCoV2 fasta file^{6 (link)}. CellBender remove-background^{75 (link)} was run to remove ambient RNA and other technical artifacts from the count matrices. The workflow is available publicly as cellbender/remove-background (snapshot 11) and documented on the CellBender GitHub repository as v0.2.0: https://github.com/broadinstitute/CellBender.